Sepsis is a common and devastating syndrome induced by infection and results from an uncontrolled inflammatory response. As highlighted in the updated consensus definition for sepsis published this year (JAMA 23; 315:801), the initial inflammatory response to infection is beneficial; but this protective response can become dysregulated and harmful. The theoretical tipping-point at which inflammation transitions from defending to destroying homeostasis is not predictable, but it identifies the point at which the risks for sepsis-related organ failure and mortality increase significantly. Consequently, improving sepsis survival requires understanding the neuro-immune interactions that define this transition, developing indices to identify this point, and testing novel therapeutics to tip the system back towards a beneficial response. The proposed project addresses each of these important knowledge gaps. Our general hypothesis is that a destructive loop exists such that peripheral inflammation evokes brainstem inflammation during sepsis, which is permissive for immune dysregulation and organ dysfunction; ultimately leading to further peripheral inflammation. In support of this hypothesis, we discovered that brainstem inflammation develops progressively in cardiorespiratory control nuclei in a rodent model of sepsis; leading to decreased efficacy of sensory feedback, reduced heart rate and ventilatory pattern variabilities, and uncoupling of autonomic and respiratory rhythms. Inhibiting brainstem cytokine expression reversed many of these changes. Furthermore, our preliminary results from a clinical trial involving patients with vasopressor-dependent septic shock identified that cardiac beat-to-beat dynamics predicted 28-day mortality with higher accuracy than standard clinical severity of illness scores. These findings underlie our specific hypotheses: 1) brainstem inflammation itself causes the loss of regulation of autonomic homeostasis and propagates the disordered inflammatory response during sepsis; and 2) cardiorespiratory uncoupling and the appearance of unstable patterns define a tipping point to a dysregulated host immune response that precedes and promotes the development of organ failure. We will test these hypotheses in an animal model of E. coli sepsis. We also propose a pre-clinical, pilot study of vagal nerve stimulation (VNS), applied as the inflammatory response tips from beneficial to harmful, as a novel `electroceutical' therapy to modulate the neuro-inflammatory response. Our planned experiments will establish a novel pathway for sepsis progression and identify markers to guide the use of VNS to oppose disease progression and improve outcomes in sepsis.